LORENTZIAN ESTIMATION FOR RICIAN DE-NOISING IN MRI IMAGES

Abstract

The accuracy of neuro imaging techniques like MRI exclusively depends in the visual analysis of an image.
Noise in the MRI is modeled as Rician distribution introduces a signal dependent bias thereby reducing the image
contrast. In this paper an efficient algorithm is proposed for Rician de-noising using Adaptive Lorentzian estimation
from the most similar neighborhood pixel extracted using Rank Ordered Absolute Difference Statistic( ROAD)statistic
after noise estimation. The performance of the algorithm is evaluated quantitatively using PSNR and by visual analysis.
The results show the efficiency of the algorithm for noise reduction and radiological analysis.